1. 54
  1.  

  2. 13

    Big fan of Firefox, have used it for years as my primary browser. Just a few days ago, I updated to 69 will happily piggyback this v70 announcement.

    In case anyone of you uses a user stylesheet on Firefox and wonders why stuff doesn’t work in FF 69 anymore, the reason is that they disabled reading user stylesheets by default. You can enable it again by setting toolkit.legacyUserProfileCustomizations.stylesheets to true.

    Don’t go down the same rabbit hole like me and try to figure out why the one thing you did in the user stylesheet doesn’t work anymore. You’ll find zillions of people with opinions - and none of their proposed solutions will work(;

    1. 4

      I hope that you have telemetry enabled. When a feature becomes not the default, that means that it is on the cutting block. If you’re not reporting that you are using a user stylesheet, then don’t be surprised when it is removed a few versions down the line.

      1. 3

        Thank you! This was driving me crazy!

        1. 2

          can you elaborate? Are you talking about userContent.css, or Stylus and similar?

          1. 1

            Sorry for the late answer. I was talking about userContent.css.

        2. 17

          I’m sad that every thread about Firefox turns into a complain-fest. It is a 28 million line product. It is hard to please everyone. Accept that it is not perfect and please don’t feel entitled that your bug must be catered to. It is a product with 100s of millions of users, hundreds of developers and dozens of teams. If it were simple to fix all the bugs and satisfy every request, it would be done. Really!

          1. 3

            I love constructive feedback and productive discussions in bugs. Please.

            1. 1

              I’m sad that every thread about Firefox turns into a complain-fest. It is a 28 million line product.

              and you jump straight to criticism too

              1. 6

                Yes I criticize the tone of the discussions.

                1. 6

                  I think the joke was that 28 millions lines of code is a ton of code for a web browser :)

            2. 5

              Friendly warning to anyone manually updating, Firefox will trash any settings you have regarding updating. So if your machine is set for manual updates, and you install Firefox 70, it changes to automatic updates. Mozilla refuses to fix this:

              https://bugzilla.mozilla.org/show_bug.cgi?id=1576400

              1. 22

                Firefox works fine when managed by, for example, a Linux distribution’s package manager. This necessarily means that the auto updater is disabled. So this use case is handled fine. It’s also open source and Mozilla is non-profit, so I’m not sure why you’ve chosen Firefox as your evil straw man. Your bug report is not productive and your childish antics have only wasted developers’ time, who have plenty of actual work to do.

                1. 2

                  I think the bug report is legitimate, as are the explanation, resolution and workaround that have been provided. You reply seems unnecessarily harsh to me, though I agree the submitter accepting the Firefox response would be appropriate.

                2. 18

                  Tbh your behaviour on that bug wasn’t really acceptable. When a maintainer closes an issue please don’t re-open it simply because you disagree. As for the “bug” itself, you were given two possible workarounds. Yet for some reason you still expect developers to spend their time catering to your very obscure edge case.

                  1. 9

                    Thats interesting framing.

                    They broke the feature. Its not me asking for something new. Its me asking them to restore the previously working behavior. Before, if you wanted to permanently disable updates, you just go into about:config and toggle, and done. Now you have to manually create a JSON file, a file which is removed the next time you manually update. Just because you call it “very obscure edge case”, doesnt make it so.

                    I imagine, especially in Lobster community, and especially in the current environment of well deserved distrust with tech companies, that its not as obscure as you think or hope it is.

                    LOL just checked your bio:

                    Engineering productivity at Mozilla

                    https://lobste.rs/u/ahal

                    1. 10

                      There is an XKCD which is precisely applicable to your situation.

                      Your use case is an edge case as far as the Mozilla developers are concerned & since it’s a tiny, tiny minority of users that are relying on it (i.e., you) then the rest of the userbase comes first. Consider yourself fortunate that the mozilla developers were kind enough to tell you how to do what you want to do in future.

                      This is the price of using a piece of software with millions of users in a unique fashion I’m afraid - your use case is never going to trump those millions of users.

                      1. 11

                        You seem to be under the impression that there’s a smoky back room at Mozilla HQ where they twirl their mustaches and cackle about how they’re deliberately taking away your freedom. Admittedly it’s been some years since I worked for Mozilla, and I was A) remote employee and B) didn’t work on the browser, but if there is such a room I certainly never heard about it, let alone got invited to go see it.

                        And what I think from reading that bug report is that you used a feature (the distribution directory) for a purpose it wasn’t intended for, and then were unhappy when it behaved as documented, because it turned out not to support the use case you want. Your use case is apparently very important to you personally, but that doesn’t mean it has to be important to them, or that they have to support it. They have the freedom to decide not to support your use case; you have the freedom not to use their software. And in fact you’re better off than you’d be with some browsers, because you also have the freedom to grab the source, modify it to suit your use case, and use and distribute your fork. But you don’t and never will have a right – moral, legal, or otherwise – to force them to support your use case.

                        1. 4

                          You seem to be under the impression that there’s a smoky back room at Mozilla HQ where they twirl their mustaches and cackle about how they’re deliberately taking away your freedom.

                          That is a strawman.

                          Changes like this always have a reason. Usually, someone runs a study, or reviews a retrospective, and finds that like 20% (I don’t know the number, but I’m sure I could find it if I looked) of Firefox users had auto-updates disabled by some adware installer or whatever. And the only way Mozilla knows of preventing other software running on the same Windows machine from changing a setting is to hardcode it into the executable, where the Windows code integrity system will ensure it doesn’t get changed.

                          That doesn’t change the fact that the solution here is removing power from the end user in ways that are frequently quite harmful. I’m pointing directly at Page Translator here. That kind of “collateral damage” is extremely messed up.

                          Hard-coding auto-updates into the EXE probably isn’t that bad (running an outdated browser with known CVEs on the Internet is just stupid). Neither is the whole practice of shipping a blocklist (obviously, allowing the blocklist to be disabled in about:config, where adware installers can change it, would completely defeat the purpose of the blocklist). The fact that it ended with an add-on that clearly isn’t adware getting blocked, on the other hand, is a scandal.

                          1. 7

                            I read the Bugzilla bug linked from that post, and it appears that there was a policy change from “side-loaded extensions can execute remote-source code” to “they can’t”.

                            I have a hard time seeing that as “a scandal”. Especially given how many times we’ve seen the pattern of an extension/add-on that used to be safe and gets taken over by an entity who abuses the extension’s privileges to do malicious things.

                            So it seems there’s been a decision that nobody gets trusted to execute remote code from an add-on, and while there are certainly going to be examples like the translator add-on that intuitively feel like they should get special exceptions to that policy, special exceptions for the “good” add-on authors don’t scale.

                            Meanwhile, the “freedom” arguments almost always really boil down to demanding that someone else write software in a way that the “freedom” supporter prefers. And I don’t see any principle of software freedom which supports forcing other people to write the things you want.

                          2. 4

                            You seem to be under the impression that there’s a smoky back room at Mozilla HQ where they twirl their mustaches and cackle about how they’re deliberately taking away your freedom.

                            You missed last week’s meeting, BTW.

                            1. 3

                              We mostly talked about you and then assigned all the bugs to you. This is what happens when you miss a meeting.

                            2. [Comment from banned user removed]

                              1. 9

                                I read the bug. One of the earliest replies told you the distribution folder is intended for third-party distributions of the browser, and that a user-initiated install of stock Firefox is intended to overwrite that, on the assumption that the third-party tweaks are no longer wanted.

                                Meanwhile, you have not stated a moral or legal principle by which you have the right to demand, and they have the obligation to provide, support for your use case. If you want to convince me, you’ll need to start there.

                                1. 2

                                  If thats the case, then that means that Firefox currently has no official way to permanently disable updates via a setting.

                                  Is that really what youre arguing?

                                  1. 4

                                    You seem to think there is some principle on which they are morally or legally obligated to provide such a setting to you because you demanded it.

                                    You still have not articulated such a principle.

                                    My personal opinion is that they have neither a moral nor a legal obligation to implement what you’re demanding. And that the point of Free/open-source software is to support that position: what you have is the right to fork the source and implement the features you want, not the right to force someone else to do it for you.

                                    1. 4

                                      Since youre refusing to answer, i will assume this is your position:

                                      Firefox currently has no official way to permanently disable updates via a setting

                                      I dont need to make any moral or legal arguments. That sentence says enough. It says that Mozilla isnt interested in giving their users agency, and thats a scary thought given their history of open source advocacy.

                                      1. 10

                                        I don’t work for Mozilla (anymore) and have no earthly idea whether there actually is a built-in, supported setting, toggle or other mechanism for what you want. You seem to be convinced there isn’t, I see no reason to disagree with that, but you also seem to think it’s some sort of vitally important thing to get people to admit it, because you also seem to think it’s some sort of terrible sin Mozilla would be ashamed to fess up to.

                                        But my entire line of questioning to you is to explore by what principle you think Mozilla has an enforceable duty to you which requires them to provide it. As I’ve repeatedly said: you have the freedom to make your own fork of Firefox which behaves the way you want it to. Nobody owes you a configuration option for a feature you want, no matter what feature it is or how important you personally think it is.

                                        The fact that you can’t seem to understand that is worrying. The fact that you also conflate “freedom” with you having the ability to demand someone else do a thing for you is also worrying.

                                        1. 0

                                          You seem to be trying to get me to say something. And I am not going to say it.

                                          My point is, and has always been since this thread started:

                                          Friendly warning to anyone manually updating, Firefox will trash any settings you have regarding updating. So if your machine is set for manual updates, and you install Firefox 70, it changes to automatic updates. Mozilla refuses to fix this.

                                          anything else I dont care about. And BTW, you have failed to refute that point, instead trying to twist it into me demanding for something. Hint, I am not demanding, just putting the word out so people are informed.

                                          1. 7

                                            Hint, I am not demanding

                                            You reopened the bug twice, including after specifically being asked not to. In what world is that not demanding?

                            3. 7

                              As an industry, the risk of users being on outdated versions of software is huge. It is prudent to take measures to auto-update and ensure that users are on the latest versions of software to reduce this risk. An inconvenience to you is a huge boon to me, and the industry as a whole.

                              1. 4

                                Thats a false choice. If a user understands the consequences, and is warned loud and clear before making this type of change, they should be allowed to do so. Thats why sudo exists and UAC on Windows.

                                1. 4

                                  TBH users been trained to click “Allow” and “I Agree” until the dialogs go away.

                                  Those decisions such as enabling auto-updating to everyone are based on statistics. They are much more likely to help than to hinder. Just displaying a dialog box is not enough these days.

                                  I too would like if it was simple toggle somewhere in settings to disable updates, but Firefox is a more complex project than people realize and there is a ton of checks, balances, and teams working on different aspects of what is in essence a little virtualized operating system with a poor choice of view model for apps.

                                  I understand why you’re frustrated, I have my pet bugs too. The good thing is that you can change stuff, you can engage in constructive dialog and send a patch. And like everyone, you need to be prepared for the team who develop that app not to want your patch or feature.

                                  The good news is that even if they don’t want that feature, there is nothing stopping you from building your own build at home or forking. Still, I’d think that if you’re forking because you don’t want auto-update because you distrust big tech, then how the hell do you expect a single person to maintain security updates for a browser? I think auto update is really good and brings in a ton of fixes in.

                                  1. 2

                                    That’s a bad analogy. You should probably thinking about how the browsers that come with Windows get updated. Mozilla needs to weigh multiple issues and stakeholders here. It does not serve them well to cater to edge cases which significantly increase risk.

                                    This is also not a user “rights” issue like your language suggests. Your rights are to take the open source code and make your own build if what Mozilla provides doesn’t work for you. Your time might be better spent looking at the overall update space and lobbying for solution that gives you more of what you want while aligning to the high-level goals of Mozilla.

                                  2. 3

                                    I don’t think users end up in about:config without meaning to and accidentally turn off automatic updates. Mozilla’s response to this issue seems silly, since they should never have removed the about:config switch in the first place. At the same time, Mozilla’s decision to remove the distribution directory seems to positively affect a large number of non-technical users who would receive a custom copy of Firefox (maybe alongside another piece of software) and then try to install vanilla Firefox.

                                    Good for non-technical users, bad for corporations trying to unify rollouts of software updates. Users like @cup are probably insignificant to Mozilla in making this decision (as they represent a very small vocal minority of the Firefox userbase).

                                    1. 8

                                      I don’t think users end up in about:config without meaning to and accidentally turn off automatic updates.

                                      Third-party software, installed on the same computer, does that. A lot of the blocked add-ons have block descriptions like “overrides search behavior without user consent or control”. Because it’s adware.

                                      I’ve seen people get infected with that kind of thing; it comes bundled with another application that they installed. If it’s in about:config, then the third-party installer can just change it. If it’s hardcoded in the EXE, then the app can’t change it without re-signing Firefox, which will get their signing key revoked by Microsoft and Apple, and likely get them sued by Mozilla for trademark infringement.

                                      1. 6

                                        That’s sneaky. I guess the fact that I wasn’t aware of that goes to show that Mozilla had more of a point here than I thought they did. Thank you for informing me.

                                        I know that sandboxing on the desktop would solve this particular problem, but I’m afraid of the consequences of that on software development and particularly people learning to code. It’s pretty difficult to be exposed to real-world programming on a locked-down mobile device.

                                        1. 1

                                          I don’t think that’s a problem.

                                          • There are perfectly good ways that an operating system vendor could allow users to turn off the sandbox without allowing arbitrary applications to do that. I particularly like the Chromebook method where you use a literal jumper on the motherboard to switch it off; it’s really just a skewmorph, to make sure the human understands that they’re doing something to their computer at a low level, but it seems effective enough.

                                            Unfortunately, Mozilla’s attempt to ship an actual operating system didn’t go anywhere, so they never got the chance to implement anything like that.

                                          • Breaking out of the sandbox is really only necessary if you want to do systems programming. I love systems programming, but most software is application code, and most of that is written with sandboxed systems like web-based JavaScript and spreadsheet macros in Excel. I can write that kind of stuff on locked-down mobile devices right now.

                                      2. 1

                                        Companies trying to unify rollouts of software is part of the problem. Many major malware incidents get investigated and the findings include well meaning administrators who wished to unify and manage updates - but, failed to update in certain situations.

                                        We shouldn’t maintain the pretense that admins will get this right 100% of the time. They are people and they will fail. Their efforts are best spent elsewhere, including work to encourage devs to test compatibility, giving admins more of a stake in software acquisitions, etc.

                                    2. 5

                                      Thanks for reporting this behavior, and also thank you for pointing out the potential conflict of interest of the person who criticized you for the way you reported it. It’s really unfortunate that the Mozilla devs are optimizing for silently updating the browser and making it difficult for users to disable this behavior. If anyone is aware of a fork of Firefox that doesn’t do this, I’d love to hear about it.

                                      1. 16

                                        The initial report was fine, it’s the way they kept reopening that isn’t

                                2. 1

                                  I think I’ve been running this for a few days while accidentally upgrading. How would that happen? Have Firefox stable installed. Want to try Firefox Beta to run on a second profile. Install Beta. Only have Beta installed. Yes, a bit surprising, but it’s been rock solid, so no real complaint here, just a little o_O. This was on Windows.